The predictive value of 18F-FDG PET/CT for assessing pathological response and survival in locally advanced rectal cancer after neoadjuvant radiochemotherapy

Purpose To evaluate whether metabolic changes in the primary tumour during and after preoperative radiochemotherapy (RCT) can predict the histopathological response in patients with locally advanced rectal cancer as well as disease-free survival (DFS) and overall survival (OS). Methods Consecutive patients with cT2-4 N0-2 rectal adenocarcinoma were included. F-18-FDG PET/CTwas performed at baseline, at the end of the second week of RCT (early PET/CT) and before surgery (late PET/CT). The PET/CT results were compared with histopathological data (ypT0 N0 vs. ypT1-4 N0-2 as well as TRG1 vs. TRG2-5) and survival. Results The study included 126 patients. Among 124 patients in whom TNM classification was available, 28 (22.6 %) were ypT0 N0, and among all 126 patients, 31 (24.6 %) were TRG1. The areas under the curve of the early response index (RI) for identifying non-complete pathological response (non-cPR) were 0.74 (95 % CI 0.61 - 0.87) for ypT1-4 N0-2 patients and 0.75 (95 % CI 0.62 -0.88) for TRG2-5 patients. The optimal cut-off for differentiating patients with non-cPR and cPR was found to be a reduction of 61.2 % (83.1 % sensitivity and 65 % specificity in ypT1-4 N0-2 patients; 85.4% sensitivity and 65.2% specificity in TRG2-5 patients). The optimal cut-off for late RI could not be found. The qualitative analysis of images obtained after RCT demonstrated 81.5 % sensitivity and 61.3 % specificity in predicting TRG2-5. After a median follow-up of 68 months, the low number of patients with local/distant recurrence or who had died did not allow the value of PET/CT for predicting DFS and OS to be calculated. Conclusion The early assessment of response to RCT by F-18-FDG PET/CT can predict non-cPR allowing practical modification of preoperative treatment. Conversely, late RI is not sufficiently accurate for guiding the decision as to whether local excision or even observation is appropriate in an individual patient. Qualitative analysis of late PET/CT images is also not sensitive enough alone to rule out the presence of residual disease.